[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-together-ai-together-embeddings":3,"mdc-c3gf3d-key":37,"related-repo-together-ai-together-embeddings":525,"related-org-together-ai-together-embeddings":620},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":12,"stars":26,"repoUrl":27,"updatedAt":28,"license":29,"forks":30,"topics":31,"repo":32,"sourceUrl":35,"mdContent":36},"together-embeddings","generate embeddings and build RAG pipelines","Dense vector embeddings, semantic search, RAG pipelines, and reranking via Together AI. Generate embeddings with open-source models and rerank results behind dedicated endpoints. Reach for it whenever the user needs vector representations or retrieval quality improvements rather than direct text generation.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},"together-ai","Together AI","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Ftogether-ai.jpg","togethercomputer",[13,17,20,23],{"name":14,"slug":15,"type":16},"LLM","llm","tag",{"name":18,"slug":19,"type":16},"RAG","rag",{"name":21,"slug":22,"type":16},"AI","ai",{"name":24,"slug":25,"type":16},"Search","search",31,"https:\u002F\u002Fgithub.com\u002Ftogethercomputer\u002Fskills","2026-07-26T05:49:06.247906",null,4,[],{"repoUrl":27,"stars":26,"forks":30,"topics":33,"description":34},[],"Skills to help your coding agents use Together AI products.","https:\u002F\u002Fgithub.com\u002Ftogethercomputer\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Ftogether-embeddings","---\nname: together-embeddings\ndescription: \"Dense vector embeddings, semantic search, RAG pipelines, and reranking via Together AI. Generate embeddings with open-source models and rerank results behind dedicated endpoints. Reach for it whenever the user needs vector representations or retrieval quality improvements rather than direct text generation.\"\n---\n\n# Together Embeddings & Reranking\n\n## Overview\n\nUse this skill for semantic retrieval components:\n\n- create embeddings\n- batch embeddings\n- build retrieval or RAG pipelines\n- rerank retrieved candidates\n\nThis skill is for retrieval plumbing, not for the final language-model response itself.\n\n## When This Skill Wins\n\n- Build vector search or semantic similarity features\n- Add embedding generation to a data pipeline\n- Improve retrieval quality with reranking\n- Assemble a retrieval stage before calling a chat model\n\n## Hand Off To Another Skill\n\n- Use `together-chat-completions` for the final answer-generation step\n- Use `together-batch-inference` for very large offline embedding backfills\n- Use `together-dedicated-model-inference` when reranking requires a dedicated deployment\n\n## Quick Routing\n\n- **Embeddings API usage**\n  - Read [references\u002Fapi-reference.md](references\u002Fapi-reference.md)\n  - Start with [scripts\u002Fembed_and_rerank.py](scripts\u002Fembed_and_rerank.py) or [scripts\u002Fembed_and_rerank.ts](scripts\u002Fembed_and_rerank.ts)\n- **Semantic search (embed, store, query)**\n  - Start with [scripts\u002Fsemantic_search.py](scripts\u002Fsemantic_search.py) -- includes an in-memory vector store, cosine-similarity retrieval, and optional rerank\n- **RAG pipeline composition**\n  - Start with [scripts\u002Frag_pipeline.py](scripts\u002Frag_pipeline.py)\n- **Model selection and rerank constraints**\n  - Read [references\u002Fmodels.md](references\u002Fmodels.md)\n\n## Workflow\n\n1. Confirm that the user needs vectors or retrieval, not direct generation.\n2. Choose the embedding model and batch shape.\n3. Generate embeddings for corpus and query paths consistently.\n4. Retrieve candidates. An in-memory cosine-similarity store works for prototyping and small corpora (see `semantic_search.py`). Use a dedicated vector database for production scale.\n5. Rerank only when the extra latency and endpoint requirement are justified. When no dedicated rerank endpoint is available, cosine-similarity ranking is a reasonable fallback.\n\n## High-Signal Rules\n\n- Python scripts require the Together v2 SDK (`together>=2.0.0`). If the user is on an older version, they must upgrade first: `uv pip install --upgrade \"together>=2.0.0\"`.\n- Keep embeddings and reranking conceptually separate; rerank is a second-stage precision step.\n- Reranking in this repo assumes a dedicated endpoint. Do not promise serverless rerank unless the product changes. When no endpoint is available, fall back to cosine-similarity ranking.\n- The embedding model has a 514-token context limit. Chunk longer documents before embedding.\n- The `rag_pipeline.py` example demonstrates retrieval plus generation; treat generation as a hand-off to chat completions.\n- Preserve model consistency across indexing and querying.\n\n## Resource Map\n\n- **API details**: [references\u002Fapi-reference.md](references\u002Fapi-reference.md)\n- **Model guide**: [references\u002Fmodels.md](references\u002Fmodels.md)\n- **Python embeddings example**: [scripts\u002Fembed_and_rerank.py](scripts\u002Fembed_and_rerank.py)\n- **TypeScript embeddings example**: [scripts\u002Fembed_and_rerank.ts](scripts\u002Fembed_and_rerank.ts)\n- **Python semantic search**: [scripts\u002Fsemantic_search.py](scripts\u002Fsemantic_search.py)\n- **Python RAG pipeline**: [scripts\u002Frag_pipeline.py](scripts\u002Frag_pipeline.py)\n\n## Official Docs\n\n- [Embeddings Overview](https:\u002F\u002Fdocs.together.ai\u002Fdocs\u002Fembeddings-overview)\n- [Rerank Overview](https:\u002F\u002Fdocs.together.ai\u002Fdocs\u002Frerank-overview)\n- [Embeddings API](https:\u002F\u002Fdocs.together.ai\u002Freference\u002Fembeddings)\n- [Rerank API](https:\u002F\u002Fdocs.together.ai\u002Freference\u002Frerank)\n",{"data":38,"body":39},{"name":4,"description":6},{"type":40,"children":41},"root",[42,51,58,64,89,94,100,123,129,170,176,281,287,324,330,387,393,475,481],{"type":43,"tag":44,"props":45,"children":47},"element","h1",{"id":46},"together-embeddings-reranking",[48],{"type":49,"value":50},"text","Together Embeddings & Reranking",{"type":43,"tag":52,"props":53,"children":55},"h2",{"id":54},"overview",[56],{"type":49,"value":57},"Overview",{"type":43,"tag":59,"props":60,"children":61},"p",{},[62],{"type":49,"value":63},"Use this skill for semantic retrieval components:",{"type":43,"tag":65,"props":66,"children":67},"ul",{},[68,74,79,84],{"type":43,"tag":69,"props":70,"children":71},"li",{},[72],{"type":49,"value":73},"create embeddings",{"type":43,"tag":69,"props":75,"children":76},{},[77],{"type":49,"value":78},"batch embeddings",{"type":43,"tag":69,"props":80,"children":81},{},[82],{"type":49,"value":83},"build retrieval or RAG pipelines",{"type":43,"tag":69,"props":85,"children":86},{},[87],{"type":49,"value":88},"rerank retrieved candidates",{"type":43,"tag":59,"props":90,"children":91},{},[92],{"type":49,"value":93},"This skill is for retrieval plumbing, not for the final language-model response itself.",{"type":43,"tag":52,"props":95,"children":97},{"id":96},"when-this-skill-wins",[98],{"type":49,"value":99},"When This Skill Wins",{"type":43,"tag":65,"props":101,"children":102},{},[103,108,113,118],{"type":43,"tag":69,"props":104,"children":105},{},[106],{"type":49,"value":107},"Build vector search or semantic similarity features",{"type":43,"tag":69,"props":109,"children":110},{},[111],{"type":49,"value":112},"Add embedding generation to a data pipeline",{"type":43,"tag":69,"props":114,"children":115},{},[116],{"type":49,"value":117},"Improve retrieval quality with reranking",{"type":43,"tag":69,"props":119,"children":120},{},[121],{"type":49,"value":122},"Assemble a retrieval stage before calling a chat model",{"type":43,"tag":52,"props":124,"children":126},{"id":125},"hand-off-to-another-skill",[127],{"type":49,"value":128},"Hand Off To Another Skill",{"type":43,"tag":65,"props":130,"children":131},{},[132,146,158],{"type":43,"tag":69,"props":133,"children":134},{},[135,137,144],{"type":49,"value":136},"Use ",{"type":43,"tag":138,"props":139,"children":141},"code",{"className":140},[],[142],{"type":49,"value":143},"together-chat-completions",{"type":49,"value":145}," for the final answer-generation step",{"type":43,"tag":69,"props":147,"children":148},{},[149,150,156],{"type":49,"value":136},{"type":43,"tag":138,"props":151,"children":153},{"className":152},[],[154],{"type":49,"value":155},"together-batch-inference",{"type":49,"value":157}," for very large offline embedding backfills",{"type":43,"tag":69,"props":159,"children":160},{},[161,162,168],{"type":49,"value":136},{"type":43,"tag":138,"props":163,"children":165},{"className":164},[],[166],{"type":49,"value":167},"together-dedicated-model-inference",{"type":49,"value":169}," when reranking requires a dedicated deployment",{"type":43,"tag":52,"props":171,"children":173},{"id":172},"quick-routing",[174],{"type":49,"value":175},"Quick Routing",{"type":43,"tag":65,"props":177,"children":178},{},[179,219,241,261],{"type":43,"tag":69,"props":180,"children":181},{},[182,188],{"type":43,"tag":183,"props":184,"children":185},"strong",{},[186],{"type":49,"value":187},"Embeddings API usage",{"type":43,"tag":65,"props":189,"children":190},{},[191,202],{"type":43,"tag":69,"props":192,"children":193},{},[194,196],{"type":49,"value":195},"Read ",{"type":43,"tag":197,"props":198,"children":200},"a",{"href":199},"references\u002Fapi-reference.md",[201],{"type":49,"value":199},{"type":43,"tag":69,"props":203,"children":204},{},[205,207,212,214],{"type":49,"value":206},"Start with ",{"type":43,"tag":197,"props":208,"children":210},{"href":209},"scripts\u002Fembed_and_rerank.py",[211],{"type":49,"value":209},{"type":49,"value":213}," or ",{"type":43,"tag":197,"props":215,"children":217},{"href":216},"scripts\u002Fembed_and_rerank.ts",[218],{"type":49,"value":216},{"type":43,"tag":69,"props":220,"children":221},{},[222,227],{"type":43,"tag":183,"props":223,"children":224},{},[225],{"type":49,"value":226},"Semantic search (embed, store, query)",{"type":43,"tag":65,"props":228,"children":229},{},[230],{"type":43,"tag":69,"props":231,"children":232},{},[233,234,239],{"type":49,"value":206},{"type":43,"tag":197,"props":235,"children":237},{"href":236},"scripts\u002Fsemantic_search.py",[238],{"type":49,"value":236},{"type":49,"value":240}," -- includes an in-memory vector store, cosine-similarity retrieval, and optional rerank",{"type":43,"tag":69,"props":242,"children":243},{},[244,249],{"type":43,"tag":183,"props":245,"children":246},{},[247],{"type":49,"value":248},"RAG pipeline composition",{"type":43,"tag":65,"props":250,"children":251},{},[252],{"type":43,"tag":69,"props":253,"children":254},{},[255,256],{"type":49,"value":206},{"type":43,"tag":197,"props":257,"children":259},{"href":258},"scripts\u002Frag_pipeline.py",[260],{"type":49,"value":258},{"type":43,"tag":69,"props":262,"children":263},{},[264,269],{"type":43,"tag":183,"props":265,"children":266},{},[267],{"type":49,"value":268},"Model selection and rerank constraints",{"type":43,"tag":65,"props":270,"children":271},{},[272],{"type":43,"tag":69,"props":273,"children":274},{},[275,276],{"type":49,"value":195},{"type":43,"tag":197,"props":277,"children":279},{"href":278},"references\u002Fmodels.md",[280],{"type":49,"value":278},{"type":43,"tag":52,"props":282,"children":284},{"id":283},"workflow",[285],{"type":49,"value":286},"Workflow",{"type":43,"tag":288,"props":289,"children":290},"ol",{},[291,296,301,306,319],{"type":43,"tag":69,"props":292,"children":293},{},[294],{"type":49,"value":295},"Confirm that the user needs vectors or retrieval, not direct generation.",{"type":43,"tag":69,"props":297,"children":298},{},[299],{"type":49,"value":300},"Choose the embedding model and batch shape.",{"type":43,"tag":69,"props":302,"children":303},{},[304],{"type":49,"value":305},"Generate embeddings for corpus and query paths consistently.",{"type":43,"tag":69,"props":307,"children":308},{},[309,311,317],{"type":49,"value":310},"Retrieve candidates. An in-memory cosine-similarity store works for prototyping and small corpora (see ",{"type":43,"tag":138,"props":312,"children":314},{"className":313},[],[315],{"type":49,"value":316},"semantic_search.py",{"type":49,"value":318},"). Use a dedicated vector database for production scale.",{"type":43,"tag":69,"props":320,"children":321},{},[322],{"type":49,"value":323},"Rerank only when the extra latency and endpoint requirement are justified. When no dedicated rerank endpoint is available, cosine-similarity ranking is a reasonable fallback.",{"type":43,"tag":52,"props":325,"children":327},{"id":326},"high-signal-rules",[328],{"type":49,"value":329},"High-Signal Rules",{"type":43,"tag":65,"props":331,"children":332},{},[333,354,359,364,369,382],{"type":43,"tag":69,"props":334,"children":335},{},[336,338,344,346,352],{"type":49,"value":337},"Python scripts require the Together v2 SDK (",{"type":43,"tag":138,"props":339,"children":341},{"className":340},[],[342],{"type":49,"value":343},"together>=2.0.0",{"type":49,"value":345},"). If the user is on an older version, they must upgrade first: ",{"type":43,"tag":138,"props":347,"children":349},{"className":348},[],[350],{"type":49,"value":351},"uv pip install --upgrade \"together>=2.0.0\"",{"type":49,"value":353},".",{"type":43,"tag":69,"props":355,"children":356},{},[357],{"type":49,"value":358},"Keep embeddings and reranking conceptually separate; rerank is a second-stage precision step.",{"type":43,"tag":69,"props":360,"children":361},{},[362],{"type":49,"value":363},"Reranking in this repo assumes a dedicated endpoint. Do not promise serverless rerank unless the product changes. When no endpoint is available, fall back to cosine-similarity ranking.",{"type":43,"tag":69,"props":365,"children":366},{},[367],{"type":49,"value":368},"The embedding model has a 514-token context limit. Chunk longer documents before embedding.",{"type":43,"tag":69,"props":370,"children":371},{},[372,374,380],{"type":49,"value":373},"The ",{"type":43,"tag":138,"props":375,"children":377},{"className":376},[],[378],{"type":49,"value":379},"rag_pipeline.py",{"type":49,"value":381}," example demonstrates retrieval plus generation; treat generation as a hand-off to chat completions.",{"type":43,"tag":69,"props":383,"children":384},{},[385],{"type":49,"value":386},"Preserve model consistency across indexing and querying.",{"type":43,"tag":52,"props":388,"children":390},{"id":389},"resource-map",[391],{"type":49,"value":392},"Resource Map",{"type":43,"tag":65,"props":394,"children":395},{},[396,410,423,436,449,462],{"type":43,"tag":69,"props":397,"children":398},{},[399,404,406],{"type":43,"tag":183,"props":400,"children":401},{},[402],{"type":49,"value":403},"API details",{"type":49,"value":405},": ",{"type":43,"tag":197,"props":407,"children":408},{"href":199},[409],{"type":49,"value":199},{"type":43,"tag":69,"props":411,"children":412},{},[413,418,419],{"type":43,"tag":183,"props":414,"children":415},{},[416],{"type":49,"value":417},"Model guide",{"type":49,"value":405},{"type":43,"tag":197,"props":420,"children":421},{"href":278},[422],{"type":49,"value":278},{"type":43,"tag":69,"props":424,"children":425},{},[426,431,432],{"type":43,"tag":183,"props":427,"children":428},{},[429],{"type":49,"value":430},"Python embeddings example",{"type":49,"value":405},{"type":43,"tag":197,"props":433,"children":434},{"href":209},[435],{"type":49,"value":209},{"type":43,"tag":69,"props":437,"children":438},{},[439,444,445],{"type":43,"tag":183,"props":440,"children":441},{},[442],{"type":49,"value":443},"TypeScript embeddings example",{"type":49,"value":405},{"type":43,"tag":197,"props":446,"children":447},{"href":216},[448],{"type":49,"value":216},{"type":43,"tag":69,"props":450,"children":451},{},[452,457,458],{"type":43,"tag":183,"props":453,"children":454},{},[455],{"type":49,"value":456},"Python semantic search",{"type":49,"value":405},{"type":43,"tag":197,"props":459,"children":460},{"href":236},[461],{"type":49,"value":236},{"type":43,"tag":69,"props":463,"children":464},{},[465,470,471],{"type":43,"tag":183,"props":466,"children":467},{},[468],{"type":49,"value":469},"Python RAG pipeline",{"type":49,"value":405},{"type":43,"tag":197,"props":472,"children":473},{"href":258},[474],{"type":49,"value":258},{"type":43,"tag":52,"props":476,"children":478},{"id":477},"official-docs",[479],{"type":49,"value":480},"Official Docs",{"type":43,"tag":65,"props":482,"children":483},{},[484,495,505,515],{"type":43,"tag":69,"props":485,"children":486},{},[487],{"type":43,"tag":197,"props":488,"children":492},{"href":489,"rel":490},"https:\u002F\u002Fdocs.together.ai\u002Fdocs\u002Fembeddings-overview",[491],"nofollow",[493],{"type":49,"value":494},"Embeddings Overview",{"type":43,"tag":69,"props":496,"children":497},{},[498],{"type":43,"tag":197,"props":499,"children":502},{"href":500,"rel":501},"https:\u002F\u002Fdocs.together.ai\u002Fdocs\u002Frerank-overview",[491],[503],{"type":49,"value":504},"Rerank Overview",{"type":43,"tag":69,"props":506,"children":507},{},[508],{"type":43,"tag":197,"props":509,"children":512},{"href":510,"rel":511},"https:\u002F\u002Fdocs.together.ai\u002Freference\u002Fembeddings",[491],[513],{"type":49,"value":514},"Embeddings API",{"type":43,"tag":69,"props":516,"children":517},{},[518],{"type":43,"tag":197,"props":519,"children":522},{"href":520,"rel":521},"https:\u002F\u002Fdocs.together.ai\u002Freference\u002Frerank",[491],[523],{"type":49,"value":524},"Rerank API",{"items":526,"total":619},[527,546,559,570,589,598,605],{"slug":528,"name":528,"fn":529,"description":530,"org":531,"tags":532,"stars":26,"repoUrl":27,"updatedAt":545},"together-audio","process audio with Together AI","Text-to-speech and speech-to-text via Together AI, including REST, streaming, and realtime WebSocket TTS, plus transcription, translation, diarization, timestamps, and live STT. Reach for it whenever the user needs audio in or audio out on Together AI rather than chat generation, image or video creation, or model training.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[533,536,539,542],{"name":534,"slug":535,"type":16},"Audio","audio",{"name":537,"slug":538,"type":16},"Speech","speech",{"name":540,"slug":541,"type":16},"Text-to-Speech","text-to-speech",{"name":543,"slug":544,"type":16},"Transcription","transcription","2026-07-26T05:49:08.246858",{"slug":155,"name":155,"fn":547,"description":548,"org":549,"tags":550,"stars":26,"repoUrl":27,"updatedAt":558},"run asynchronous batch inference jobs","High-volume, asynchronous offline inference at up to 50% lower cost via Together AI's Batch API. Prepare JSONL inputs, upload files, create jobs, poll status, and download outputs. Reach for it whenever the user needs non-interactive bulk inference rather than real-time chat or evaluation jobs.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[551,554,555],{"name":552,"slug":553,"type":16},"Automation","automation",{"name":14,"slug":15,"type":16},{"name":556,"slug":557,"type":16},"Performance","performance","2026-07-17T06:08:23.919602",{"slug":143,"name":143,"fn":560,"description":561,"org":562,"tags":563,"stars":26,"repoUrl":27,"updatedAt":569},"generate text with Together AI","Real-time and streaming text generation via Together AI's OpenAI-compatible chat\u002Fcompletions API, including multi-turn conversations, tool and function calling, structured JSON outputs, and reasoning models. Reach for it whenever the user wants to build or debug text generation on Together AI, unless they specifically need batch jobs, embeddings, fine-tuning, dedicated endpoints, dedicated containers, or GPU clusters.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[564,565,568],{"name":21,"slug":22,"type":16},{"name":566,"slug":567,"type":16},"API Development","api-development",{"name":14,"slug":15,"type":16},"2026-07-26T05:49:11.244104",{"slug":571,"name":571,"fn":572,"description":573,"org":574,"tags":575,"stars":26,"repoUrl":27,"updatedAt":588},"together-dedicated-containers","deploy custom inference containers on Together AI","Custom Dockerized inference workers on Together AI's managed GPU infrastructure. Build with Sprocket SDK, configure with Jig CLI, submit async queue jobs, and poll results. Reach for it whenever the user needs container-level control rather than a standard model endpoint or raw cluster.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[576,579,582,585],{"name":577,"slug":578,"type":16},"AI Infrastructure","ai-infrastructure",{"name":580,"slug":581,"type":16},"Deployment","deployment",{"name":583,"slug":584,"type":16},"Docker","docker",{"name":586,"slug":587,"type":16},"Machine Learning","machine-learning","2026-07-26T05:49:09.267892",{"slug":167,"name":167,"fn":590,"description":591,"org":592,"tags":593,"stars":26,"repoUrl":27,"updatedAt":597},"deploy and operate Together AI models","Deploy and operate models on dedicated GPUs with Together AI's Dedicated Model Inference (DMI, the v2 dedicated endpoints API): beta endpoints, deployments, deployment profiles and hardware configs, autoscaling, traffic splitting, A\u002FB tests, shadow experiments, Prometheus metrics, and custom model or LoRA adapter uploads. Reach for it whenever the user mentions together beta endpoints or tg beta commands, client.beta.endpoints, DMI resources like ep_\u002Fdep_\u002Fcr_\u002Fml_ IDs, or wants production model serving with traffic management on Together AI. This is the current dedicated-hosting API and also covers migrating off the retired legacy v1 endpoints API (non-beta client.endpoints \u002F together endpoints), whose create and restart now return HTTP 403.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[594,595,596],{"name":577,"slug":578,"type":16},{"name":580,"slug":581,"type":16},{"name":14,"slug":15,"type":16},"2026-07-26T06:08:44.044039",{"slug":4,"name":4,"fn":5,"description":6,"org":599,"tags":600,"stars":26,"repoUrl":27,"updatedAt":28},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[601,602,603,604],{"name":21,"slug":22,"type":16},{"name":14,"slug":15,"type":16},{"name":18,"slug":19,"type":16},{"name":24,"slug":25,"type":16},{"slug":606,"name":606,"fn":607,"description":608,"org":609,"tags":610,"stars":26,"repoUrl":27,"updatedAt":618},"together-evaluations","evaluate LLM outputs with Together AI","LLM-as-a-judge evaluation framework on Together AI. Classify, score, and compare model outputs, select judge models, use external-provider judges or targets, poll results and download reports. Reach for it whenever the user wants to benchmark outputs, grade responses, compare A\u002FB variants, or operationalize automated evaluations.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[611,614,617],{"name":612,"slug":613,"type":16},"Benchmarking","benchmarking",{"name":615,"slug":616,"type":16},"Evals","evals",{"name":14,"slug":15,"type":16},"2026-07-26T05:49:07.241553",14,{"items":621,"total":744},[622,643,655,662,668,674,681,687,694,700,710,725],{"slug":623,"name":623,"fn":624,"description":625,"org":626,"tags":627,"stars":640,"repoUrl":641,"updatedAt":642},"blog-post","write structured blog posts","Write structured long-form blog posts with a consistent structure and SEO optimization.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[628,631,634,637],{"name":629,"slug":630,"type":16},"Content Creation","content-creation",{"name":632,"slug":633,"type":16},"Marketing","marketing",{"name":635,"slug":636,"type":16},"SEO","seo",{"name":638,"slug":639,"type":16},"Writing","writing",1151,"https:\u002F\u002Fgithub.com\u002Ftogethercomputer\u002Ftogether-cookbook","2026-07-17T06:07:39.475433",{"slug":644,"name":644,"fn":645,"description":646,"org":647,"tags":648,"stars":640,"repoUrl":641,"updatedAt":654},"social-media","create social media content","Create social media content including Twitter\u002FX threads, LinkedIn posts, and short-form updates.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[649,650,651,653],{"name":629,"slug":630,"type":16},{"name":632,"slug":633,"type":16},{"name":652,"slug":644,"type":16},"Social Media",{"name":638,"slug":639,"type":16},"2026-07-17T06:07:38.455058",{"slug":528,"name":528,"fn":529,"description":530,"org":656,"tags":657,"stars":26,"repoUrl":27,"updatedAt":545},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[658,659,660,661],{"name":534,"slug":535,"type":16},{"name":537,"slug":538,"type":16},{"name":540,"slug":541,"type":16},{"name":543,"slug":544,"type":16},{"slug":155,"name":155,"fn":547,"description":548,"org":663,"tags":664,"stars":26,"repoUrl":27,"updatedAt":558},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[665,666,667],{"name":552,"slug":553,"type":16},{"name":14,"slug":15,"type":16},{"name":556,"slug":557,"type":16},{"slug":143,"name":143,"fn":560,"description":561,"org":669,"tags":670,"stars":26,"repoUrl":27,"updatedAt":569},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[671,672,673],{"name":21,"slug":22,"type":16},{"name":566,"slug":567,"type":16},{"name":14,"slug":15,"type":16},{"slug":571,"name":571,"fn":572,"description":573,"org":675,"tags":676,"stars":26,"repoUrl":27,"updatedAt":588},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[677,678,679,680],{"name":577,"slug":578,"type":16},{"name":580,"slug":581,"type":16},{"name":583,"slug":584,"type":16},{"name":586,"slug":587,"type":16},{"slug":167,"name":167,"fn":590,"description":591,"org":682,"tags":683,"stars":26,"repoUrl":27,"updatedAt":597},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[684,685,686],{"name":577,"slug":578,"type":16},{"name":580,"slug":581,"type":16},{"name":14,"slug":15,"type":16},{"slug":4,"name":4,"fn":5,"description":6,"org":688,"tags":689,"stars":26,"repoUrl":27,"updatedAt":28},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[690,691,692,693],{"name":21,"slug":22,"type":16},{"name":14,"slug":15,"type":16},{"name":18,"slug":19,"type":16},{"name":24,"slug":25,"type":16},{"slug":606,"name":606,"fn":607,"description":608,"org":695,"tags":696,"stars":26,"repoUrl":27,"updatedAt":618},{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[697,698,699],{"name":612,"slug":613,"type":16},{"name":615,"slug":616,"type":16},{"name":14,"slug":15,"type":16},{"slug":701,"name":701,"fn":702,"description":703,"org":704,"tags":705,"stars":26,"repoUrl":27,"updatedAt":709},"together-fine-tuning","fine-tune and adapt models on Together AI","LoRA, full fine-tuning, DPO preference tuning, VLM training, function-calling tuning, reasoning tuning, and BYOM uploads on Together AI. Reach for it whenever the user wants to adapt a model on custom data rather than only run inference, evaluate outputs, or host an existing model.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[706,707,708],{"name":21,"slug":22,"type":16},{"name":14,"slug":15,"type":16},{"name":586,"slug":587,"type":16},"2026-07-26T05:49:10.243114",{"slug":711,"name":711,"fn":712,"description":713,"org":714,"tags":715,"stars":26,"repoUrl":27,"updatedAt":724},"together-gpu-clusters","orchestrate GPU clusters on Together AI","On-demand and reserved GPU clusters (H100, H200, B200) on Together AI with Kubernetes or Slurm orchestration, shared storage, credential management, and cluster scaling for ML and HPC jobs. Reach for it when the user needs multi-node compute or infrastructure control rather than a managed model endpoint.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[716,717,720,723],{"name":577,"slug":578,"type":16},{"name":718,"slug":719,"type":16},"Cloud","cloud",{"name":721,"slug":722,"type":16},"Kubernetes","kubernetes",{"name":586,"slug":587,"type":16},"2026-07-26T05:49:05.252646",{"slug":726,"name":726,"fn":727,"description":728,"org":729,"tags":730,"stars":26,"repoUrl":27,"updatedAt":743},"together-images","generate and edit images with Together AI","Text-to-image generation and image editing via Together AI, including FLUX and Kontext models, LoRA-based styling, reference-image guidance, and local image downloads. Reach for it whenever the user wants to generate or edit images on Together AI rather than create videos or build text-only chat applications.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":11},[731,734,737,740],{"name":732,"slug":733,"type":16},"Creative","creative",{"name":735,"slug":736,"type":16},"Design","design",{"name":738,"slug":739,"type":16},"Image Generation","image-generation",{"name":741,"slug":742,"type":16},"Multimodal","multimodal","2026-07-17T06:04:23.067755",16]